Governing freedom

When the most influential people in the internet are assembled, Esther Dyson never fails to appear high up the list. She is best known as a commentator, thought leader and networker, through her industry newsletter Release 1.0 and her book, Release 2.0: a design for living in the digital age.

Dyson played a key role in building internet governance. She was the first chair of ICANN (the Internet Corporation for Assigned Names and Numbers), the international body with responsibility for the internet’s domain name and addressing systems. She also made a pioneering contribution to online civil liberties with the Electronic Frontier Foundation.

Here Dyson tells the inside story of how ICANN came into being and the path she has tried to navigate between big government, corporate business and technical professionals. She presents her vision for opening up the web to the scrutiny of its far-flung community of users. In the 1990s, she contended that the internet could dissolve party politics. But now, in openDemocracy, she argues for global political parties of the internet itself.

Is this the best approach? We invite you to continue the interview by posing your responses to Esther Dyson.

openDemocracy – In Release 2.0 you argue that the impact of the internet and the web may not change human beings – we will remain hungry, lazy, creative, hateful and loving. But the institutions we’ve inherited will be transformed, if not replaced, by the impact of the new technology. One of these institutions is the nation state. As if to demonstrate its decline, you helped create a new institution of international governance uniquely free of nation states. You participated in the birth of the internet, not just as an American but also in Russia.

Esther Dyson – I wasn’t living in Russia, but I visited a lot. I watched the moment of connection when the net suddenly took life in Russia. This was after 1989. It wasn’t really the internet yet. It wasn’t online. It was just sending files from one machine to another. The European Academic Research Network gave them free access through Helsinki and the traffic exploded.

open – The way the government of the internet evolved seems genuinely novel. ICANN (the internet Corporation for Assigned Names and Numbers) has developed outside the usual international structures, without calling in the United Nations…

ED – though they certainly offered. Also, ICANN is not yet finished, so don’t speak too soon. And it is not a government in the traditional sense. It has no statutory authority. Its only power is through contracts with the entities it regulates. They don’t have to sign these contracts although they don’t get access to the facilities that ICANN oversees – the domain names, for one – unless they do.

open – Could you tell us how it started. Was there a political decision that led to it?

ED – ICANN was created at the behest of, but not by, Ira Magaziner, in the Clinton administration. He had a very enlightened attitude. He argued that running the internet and its infrastructure is not a job for the US government, or any government. He concluded that it was a job for the internet community itself, and that it should create its own governance structure.

open – Was there another example he based this on?

ED – No. It was an attempt to foster immaculate conception.

open – Where did he get that idea from?

ED – He had been working with Hillary Clinton to help reform health care in the USA. To put it kindly, their effort was not very successful. As I understand it, Bill said to him, OK Ira, why don’t you find yourself another project – because they were very close. And Ira, whom I had met long before when he was a business consultant doing some work with Wang, found the internet. I had been a sub-contractor to him. We worked together briefly for Wang, to no great good for poor old Wang. Then years later, perhaps because of that, or because I was already known in this thing, he created a very informal advisory group. I think I still have his invitation in an email folder. I called it the “Magaziner Club”.

Ira set out a global framework for e-commerce. It was a first draft and we looked at it. There were maybe twelve points, and one had to do with privacy, another was freedom of speech. I remember sitting around in his office (this would be around 1996). The draft said that the administration and technical infrastructure of the internet should be managed by the internet community itself, in some form or other, and should not be regulated by the US government.

I wasn’t terribly technical. So to me this wasn’t of great interest. I was more concerned about privacy. But when this paper came out, it caused quite an uproar in various quarters. The privacy stuff upset people in Europe. The proposals on the internet administration caused a real uproar among the people who were currently doing it. This was mainly a company called Network Solutions (NSI), which had a government-granted monopoly contract.

A central guy in it all was Jon Postel. Jon had a beard and hair down to the top of his behind, and Ira was so proud to have taken him to lunch at the White House. He was out at the university of Southern California, part of a group who were working on government contracts to do various stuff including the DNS (domain name system) root server. Generally, they were the people who defined the technical protocols of the early internet. They produced the RFCs, the Requests for Comments, which were defacto standards and policies. He had been doing this single-handedly and it was getting to be too much of a workload; he had a bunch of pending requests for new TLDs - the top level domain names.

Network Solutions, of course, didn’t want to give up the contract. So pretty soon there were two big camps – the Jon Postel camp and the NSI camp. NSI was knee-jerk, supported by the business people. And Jon was knee-jerk, supported by the technical community.

Ira had these notions of democracy. Both the two sides pretty much thought this was crazy. The engineering people said, ours is a holy trust and we are the priests. Just let us keep doing it; don’t you guys interfere. The established business community said, this thing works; don’t screw around with it. NSI may be a monopolist, but it’s working. Leave it alone, and for god’s sake don’t create new TLDs.

Almost nobody was in favour. Congresspeople who noticed Ira’s proposal thought it was a plot to give away an American birthright. The European Union said, in effect, you’re nuts, but if you don’t want it, we’ll take it. That’s not a quote, but that was by all accounts their attitude. Most other governments in the world didn’t know what was going on. If they did, they thought it sounded like some kind of crazy American democratic idealist sort of weird thing. Meanwhile the American left thought it was a plot to take something that was run by government for all the people and put it into private hands.

Once it was seen to be happening, some people did start to say, well let’s try and do this right. But there were lots of flame wars, documents, meetings and hearings about how it should actually happen and who should end up being on the board. It was really confusing. In my book, I pointed to it as an example of internet governance emerging. I optimistically said I was sure it was going to work OK because I couldn’t figure out what was going on!

When they finally decided to pick a board, they were looking for people who were not on one side or the other of the techie/business divide. So, for my sins, I got picked as someone who wasn’t partisan. The way it actually happened, I was in Aspen in the summer of ’98 for a conference. Ira Magaziner was there and also Roger Cochetti. Both of them came up to me separately and said, I’m not making a request, but if they were to create a board to run this thing, and if you were to be invited to sit on it, do you think you would be likely to accept? The wording was very carefully pitched. I said, yeah, it sounds kind of interesting. I’m always up for adventure.

I thought I was going to get an offer in about a week or two, but nothing. Much later I got an email from a guy called Joe Sims saying they were creating this board and would I like to sit on it. I’d never heard of him, so I e-mailed Cochetti and asked who is this Joe Sims guy, is he for real or some kind of hoax? He said, oh that’s Jon Postel’s lawyer. Yes, it’s for real.

But before the board could actually meet, on 25 October 1998, Jon Postel died following heart surgery. Even though this is a conspiracy-laden field, no serious person ever suspected foul play. It was a really unfortunate coincidence, because the whole thing began with no moral authority, no presiding saint. Today people talk about Ira Magaziner’s White Papers (the final draft of his framework) as if it were the Holy Scripture, and they talk about Jon Postel as the saint who went to heaven. And each person claims that if Jon were still alive, he’d have done exactly what they themselves want. Of course, if Jon Postel were really alive, he would have changed his mind about many things, just like a normal human being. Dead saints are one of the most pernicious things in the world!

Though not saintly, Jon was wise and good at getting people to resolve conflicts, and totally without vested interest in anything other than the public good. But there he was being represented in his after-life by a corporate anti-trust lawyer. The previous February, Jon had had an issue over something that the government was doing and he had redirected the root server. The University of Southern California told him to get his own lawyer because they were not responsible. And so he had gone and gotten Joe Sims.

Joe is a great, smart guy. He wants to do the right things, but he sometimes seems to think people are interfering, rather than that maybe they have a point of view that should be listened to and respected. In fact that was the flavour of the entire ICANN board – we’re right and don’t criticise us! Most of us were not used to public scrutiny. They were well-meaning, but they didn’t know much about the internet and didn’t use email very well.

They were picked not to be partisan. But, to be honest, I think they made a mistake picking these people – not the people in particular, but in the overall approach.

Creating the ‘thing’

open – Who was the ‘they’ doing the picking?

ED – As far as I know, it was basically Jon. He had consulted everybody, including the EU. They made much more of a mystery of it than they should have. It was a real problem because it was in fact a very closed process. As I said, I heard nothing until suddenly I got an email. They argued they could not go around waving people’s names in the air and getting public approval.

open – Where did the power come from which allowed Jon Postel and Joe Sims to do this?

ED – What Ira said was, the US government has this authority. We can’t figure out where it came from. But the fact is, these things exist and they’re being run under contract to the US government, and we don’t think that’s right. We think the internet community itself should create a supervisory body that is legitimate and reflects the will of the internet community. Then, he argued, the government has enough authority to decide to hand its power over. Ira was saying that the government did not feel it was the legitimate holder of this power, and that it should belong to the internet community and that when the internet community gets together and creates something that is legitimate, he would pass authority over to it.

And so this ‘thing’ was created. It may well have been as legitimate as anything could be, but it certainly didn’t have widespread trust and approval. Then there was the negotiation with Network Solutions. It had a monopoly and knew that ICANN was being created to destroy this monopoly, and so it fought tooth and nail, which is why you needed a tough guy like Joe Sims on the other side. He was very necessary and he did the right things, but he bruised some feelings along the way.

open – Where did the Network Solutions monopoly come from?

ED – They had a government contract. It was completely legal. They originally had a contract to maintain the name database for free. It was another government service – your taxpayer dollars at work. And of course the irony was that the US government was doing this for anybody in the world, free. You’d write to Network Solutions and you’d get a domain name and .com, even if you were somebody from outside the US who wasn’t paying US taxes. And so of course it wasn’t appropriate long term. The business got bigger and more costly for the government to fund, so the government said, you can charge the users a little more than just to cover your costs, and it’s a contract we can negotiate. They gave them, I guess, a very long and ultimately more lucrative contract than expected. The price for a domain name was seventy dollars or something. NSI didn’t earn this in the way Microsoft earned whatever monopoly they got. But on the other hand, they didn’t have to do anything nasty to get it. Which is the worst monopoly – the one you got for nothing, because it was granted legitimately, or the one you got because you stomped on your competition? Both are pernicious if abused.

Opening the doors

That was the situation, and the new ICANN board walked into an atmosphere of mistrust. They believed they were doing the right thing, and they were told they were doing the right thing. Most didn’t know the internet community. So then there followed negotiations between us – I was chairman – and Ira. Ira wanted the system to be more open and visible and accountable than it was. Both Joe Sims and, as far as I can recall, NSI, were saying this is ridiculous, you can’t expect these people to have public meetings. (Being idealistic, I had said we should have open meetings.) But I wasn’t telling the board what to do. My job was only to express the board’s opinion as a whole. Unfortunately it had very few opinions, because it didn’t know very much, and its opinions were supposed to reflect a mostly nonexistent consensus.

The first meeting was in Boston. I said, OK, if we can’t have an open board meeting, let’s have a public meeting beforehand so at least they can see who we are and watch us think. It was really nasty. The board started seeing enemies and thinking, this is even worse than we thought.

open – Nasty in what way?

ED – Well, people were saying, who the hell are you? You can see the transcript. It was like – who appointed you, and isn’t Joe Sims a liar, and isn’t Mike Roberts a stooge, and isn’t Esther Dyson in the pay of IBM? The board finally responded to the bad press and ultimately made our board meetings open. Of course it was a complete disappointment, because the whole board talked very politely and voted in unison, except that I voted against one thing on disclosure of conflict of interest. (I said we should disclose what the conflicts were to the public rather than simply to a committee.) But people were hoping to see a real board in action and of course they didn’t. Then one of our members, at the next meeting in Santiago, Chile, told a journalist that everything was decided over dinner the night before! And he was the one who wanted them closed.

It was all very legal and clever, but the problem was that it didn’t generate the trust it needed. Here was ICANN, but there was no civil society to support it.

open – How was all this funded?

ED – Good question. The theory was that we would be appointed and that we would immediately negotiate a contract with Network Solutions and we’d get a small piece of their fees. And then we got put into a box. Maybe it was NSI being clever; maybe it was just some well-meaning outsiders, who knows? Anyway the notion began to float around that ICANN was taxing the Internet.

The following summer we had a congressional hearing. I was happy to do this, because I wanted to know what it was really like. I testified in a hostile hearing of the Subcommittee on Oversight and Investigations.

We were still in the midst of our fight with NSI. We were being accused of taxing the internet and of destroying a god-fearing American business. Joe Sims prepped me. This is what you do before any testimony or hearing in a lawsuit. He asked me questions, and I answered them. He said, just answer the question, don’t give long explanations, if they want more they’ll ask you. Just answer the question politely and shut up. He’s really good, and I like him, he’s smart, but he’s not ‘culturally sensitive’.

I went in there and these guys were really gunning for us. They suggested we were trying to screw up the internet, that we were tools of the Clinton Democrat administration, and we were taxing Americans. I explained how we were created, and that the idea was to limit monopoly power and create competition and lower the prices for people buying domain names and so forth.

There’s two parts of the business. There’s the registry: maintaining the database; and then there’s the registrars’ services: dealing with the customer. It’s wholesale and retail (and at this point we had already gone through the process of appointing new registrars to compete with NSI). The guy representing NSI was its new CEO, Jim Rutt. It seems he had not been prepped by a lawyer. He also didn’t know that you’re not supposed to address Senators by their first name, or contradict them. They asked him some questions, and he said ICANN is making all these decisions and people’s voices are not being heard. One of the Senators said, well for example Mr Rutt, whose voices are not being heard? Well people’s voices are just not being heard, and ICANN… Excuse me Mr Rutt, whose voices are not being heard? And he said, well I’ll have to get back to you on that.

There are conspiracy theories about everything. Someone came up to me afterwards jokingly and said, so how did you get Network Solutions to hire this guy? After that things got much easier.

open – This means that without the Clinton Administration…

ED – …it probably would not have happened. It was an odd example of idealism for both the free market and for the internet as a separate space. The Republicans, who are more free market, may not have had the imagination to do this.

open – They would have kept the NSI monopoly?

ED – It’s a really interesting question. Ira had a peculiarly American idealism, which ran up against state authority on the one hand, and business with its tremendous powers on the other. The trademark interests wanted intellectual property to be properly managed and domain names and trademarks to be well regulated, and they didn’t want a proliferation of these new TLDs, which we now have. That is one current issue: do we have too many, or too few? Are we doing it badly? Is ICANN now a tool of these insider corporate interests and the engineers? And what about the public?

Democracy for the internet?

ICANN is different from other international organisations because any regular person with a PC, from Kansas to Kathmandu, needs good internet access and can buy a domain name themselves. People use the internet directly for themselves and are therefore directly influenced by its governance. They are its stakeholders, so to speak.

Most people say well, we’ve got it [ICANN], we’re stuck with it, let’s try and make it better – but how? One view is that the internet is a giant global democracy, therefore ICANN should be elected by every internet user or maybe every potential internet user. So let’s just give everybody a vote.

Of course, most people couldn’t care less about ICANN. So there are real dangers with such a view. What if AOL – with nearly 30 million members – encouraged them to vote for whatever?

On the other side, there are those who say, let’s keep the clueless, unwashed people out of here. We’re the engineers, the IP interests, we know how to run this thing. Leave it to the experts. This too is a plausible point of view. They are saying, we understand how this works, we don’t want to screw it up, we’re not trying to use it to control content or to dominate the world, we just want the internet to be effective and work well because then we can make money. So why don’t you leave us alone, and let us run this thing properly?

open – How, then, would you like there to be a connection with the internet public?

ED – ICANN has a sort of bicameral board. It has two sets of nine members. The first are from the so-called supporting organisations. These are from three clearly defined constituencies. There is the Protocol Supporting Organisation, which oversees the technical protocols. There is the Address Supporting Organisation, which looks after the IP addresses, the numbers. Then there is the DNSO, which is the Domain Name Supporting Organisation, which has intellectual property interests and registrars and registries, and business and ‘all other’.

An Eastern European once said to me that he wanted to start a political party but he was waiting for government funding. In essence what you have here are three parties with government funding. In my view, the public needs to get its own funding. The IP address people are actually pretty good – they are open and it’s a relatively open organisation. But of course there is not a lot of loot there. There is some: if you have a nice block of IP addresses you can sell access to them as an ISP. The protocols are very technical and one standard versus another is of economic interest to somebody or other, but it’s pretty OK, in a broad sense. On the other hand, DNSO has seven different groups and they elect three members to the Names Council and in turn they elect three directors to the ICANN board. This grants power to seven arbitrary interest groups, rather than making each interest group earn it.

The second half of the board is another nine directors who represent the at-large membership. They have to earn their position, but how can you earn it when you don’t know who your constituency is? Or suppose NTT or somebody in Japan decides to register lots of people so they can vote in a particular way. There’s some kind of ugly nationalism in there. I’m wrestling with this now, because I’m on the AtLarge Membership Study Committee. The chairman of the committee is Carl Bildt, former Prime Minister of Sweden. The question is – how do you represent the public interest? Is it the informed public? How do you determine whether they’re informed? Who gets to decide whether these people are legitimate? Should they have a test, should they have to pay £10 to show that they’re interested? Because otherwise somebody could maybe create millions of fictitious people. How do they vote? Should they vote per region? Almost all the Germans voted for the German candidate.

I would like to see parties emerge. A ‘More New TLDs Party,’ or an ‘IP is Sacred’ party…

open – World parties?

ED – Yes. They would represent an interest rather than a nationality or a region. Voting for individuals doesn’t mean much on its own. You need parties with platforms. If you have a platform you have to address the questions which follow. For example, if you’re going to give everybody free bread, how do you expect to pay for it? But if you just ask people to vote for free bread, of course they will. We need a layer of civil society in which people say, well I’m for free bread, and I propose we fund it by this bread tax.

open – You want to reverse engineer civil society into this institution?

ED – Around it, yes. It’s got the naked scaffolding. But what I would like to see is these sorts of parties emerging without government funding, without ICANN decreeing them. That’s the whole point of civil society: that it shouldn’t depend on government; it should create its own movement. This is partly why I’m hoping somebody will read this in the UK or Sweden or India and say, ‘Right – I’m going to start the New TLDs Party’.

open – You want global activity to influence an institution like ICANN, which, although it was state-sponsored and represents interest groups…

ED – It wasn’t state-sponsored. At the moment, it is its own ‘state’ but it has no statutory authority. And it has no civil society attached to it. It is so important to understand this, even though you can say it was more in the letter than in the spirit, but it was not state-sponsored. The state said, please go do it yourselves. Now the people have to do it for themselves.

open – Your experience in Washington was very complicated and difficult. Isn’t one of the worries about creating a political civil society that it’s really awfully painful?

ED – You mean, isn’t politics disgusting? Yes, but the alternative is worse, which is hidden politics. It is state or dictator control and it has no transparency and it has no messiness. The problem is, you could say that all power is illegitimate. ICANN, with its system of contracts and consensus, is an attempt to build a system of legitimate power.

A good thing about ICANN is it’s very firmly defined in terms of what it can do. Of course everybody is scared that it is going to metastasise into doing other things, like the EU, for instance. This is why it needs a large constituency to which it is actively accountable.

open – So you want the people to take power?

ED – No. You want to reduce and diffuse the power, rather than change who has it. Believe me, if you put a lot of people from outside in power, they would want to do terrible things like control the content. Beware of giving everyone the vote, because you might not like what they vote for. But likewise, with most governments – you don’t really want to deal with them either. I really want the rest of the world to be part of ICANN. But as Ira found, what happens is not what you expect. There are a lot of governments out there who want much tighter control over everything, to shut down the sites that have ‘dangerous’ content, defined as ‘dangerous to this particular state government’. Those of us who want more accountability are trying to do an end-run around these guys too. They see ICANN as a tool that can help them control people. But ICANN is designed only to control resources. That’s a fine distinction, but a key one, an important principle.

We are talking only about ICANN. But to have effective governance of any sort you need parties, institutions, media, places where people can go to. Democracy isn’t just about voting, just as the market is not just about transactions. It’s about people getting together to discuss what they want their government to do and understanding the trade-offs between various things, such as taxes and personal choices. Business, likewise, is not just transacting: it’s understanding what’s available, who’s a good person to deal with, and who’s not.

A free market must have lots of institutions in addition to economic actors and the state. This is what I mean by civil society. I’m still, if you like, right wing. But I very much appreciate the role of governments and regulation, as well as civil society. I learned this in Russia.

The Russian lesson

When I went there I began to understand what it is like when civil society, self-organising markets, don’t exist. I’d been living like a fish in water my whole life, and I suddenly recognised, ah, so that’s what water is like. It made me much more aware of the complexities. The West did not help when it waltzed in and said, let’s have a free market, without understanding how much regulation and trust a free market requires.

Put it like this: Bill Gates is not the reason for America’s success; the reason for America’s success (amongst other things) is all the users of Bill Gates’ products. The fact that Bill Gates is a billionaire is not very interesting. What is interesting is that all these people are using his products to make their businesses more effective and their communications with people better, and that they took to it so fast – that’s the amazing thing. That’s what you want to see in Russia. I had never been a fan of the outsourcing business – which I see as sell your services abroad to bring money in. But at this point I have recanted, because bringing the money in and getting people trained is better than not having work. But the real success is going to be when these Russian IT companies do most of their business in and for Russia.

open – Is that why you work there a lot?

ED – I do it because I like challenges. What you also have to acknowledge to yourself is you’re going to die before you’ve done everything you would have liked to do, so it’s simply a question of setting priorities. Don’t try and fix the world, do try and fix the parts of it that you happen to come into contact with. Calm down, you’re not going to get to the end of the list, so just start where it matters and relax, do what you can.

Everybody has their own levels of how selfish or selfless they want to be. If you get too selfless, then you end up having nothing to give. You need a self with which to create something from which you can give. If you’re too selfish you give nothing; you keep it all. You have to make a balance – judgements about what you will take on and what you will abandon to somebody else. What we’re doing with Bridges.org in South Africa, is just to be a model for other people so they can help save their bits of the world.

open – You say you don’t trust anything big, but the internet seems like the perfect tool for organisations, a) to become big, and b) to project trust onto ignorant consumers.

ED – No, TV is much better. Broadcasting is a medium for propaganda; the internet is a medium for conspiracy. The internet is always going to foster little side groups, little whispers. You can find other nutcases just like yourself, find other freedom fighters just like yourself. The internet is always about things at the edges eroding what’s at the centre. It’s a platform for little things self-organising. It’s fractal.

The whole point about the internet is not whether you’re big or small, it’s about dynamism. Whatever you are, you have to keep transforming, because if you just stay static you’ll crumble. That’s why I like it. If there could be an organisation that was big and transparent, really open, it would have too many things poking at it to stay still. It would always be in flux.

open – At one point in your book you say that in order to ride this wave of change people need emotional intelligence.

ED – I said that?

open – Yes, and you ask if it is not de-equipping people to handle the thing that is being created.

ED – I don’t know. Personally, I like turbulence in airplanes, but a lot of people find the motion really scary. People who are experienced in boats understand they’re riding the waves and they feel secure. You need a feeling of security amidst the turbulence, amidst the flashing. It may be that the internet is equipping people to think deeply despite the flashing lights. And of course there were a lot of people born a hundred years ago who never thought deeply either.

On being a good capitalist

open – As you were there at the start, we must ask what you think of the claims that the internet revolution has created a new model of capitalism that is leading to benign and perpetual growth.

ED – I don’t think there is a new model of capitalism. I think there is a new texture and there is a new balance of power. But to me, capitalism is about long-term investment. It’s not trading. It’s not mercantilism. This is what was lacking in Russia. They had transactions, but they don’t have the oxygen of trust that gives life. If you don’t have trust, you won’t invest because you don’t have faith that you’ll get a return on what you invest. The ultimate form of capitalism, for me, is philanthropy. You’re so trusting you’ll get a return – you don’t even need to own it. This is slightly too sweet. But if you draw a chart and you plot short-term versus long-term, the good capitalists think long-term, they invest for long-term returns. And the short-termers are day-traders.

The short-term philanthropists give money to beggars in the street and they feel good, or they send ten dollars to twenty-nine charities. The long-term investors are ‘capitalist’ philanthropists. They invest. They say, I’m not just going to give my money away to feel good, I’m going to invest my money, I’m going to put it into three projects and I’m going to watch what these guys do. I’m not going to expect an immediate return. I won’t feel good when I give the money. I’ll feel good when I see it working, when I see a return on the money. There’s a lot more in common in the time horizon of good capitalists and good philanthropists.

The real long-termers want a better world for their grandchildren: in terms of what their grandchildren will have, and for those their grandchildren will live among.